tellarin

Dupe: https://news.ycombinator.com/item?id=48727789

Some collaborators and I recently released a first version of a benchmark we think highlights a critical gap in recent AI models in understanding causality in the real-world, beyond a physics focus.

Everyday environments are rich in tangible control interfaces (TCIs), like, light switches, appliance panels, and embedded GUIs, that are designed for humans and demand commonsense and physics reasoning, but also causal prediction and outcome verification in time and space (e.g., delayed heating, remote lights).

Paper: SWITCH: Benchmarking Modeling and Handling of Tangible Interfaces in Long-horizon Embodied Scenarios

Data and leaderboard in HuggingFace.

Feedback, suggestions, and collaborators are very welcome!

Selfless plug here... Some collaborators and I just released a first version of a benchmark we think highlights a critical gap in recent models in understanding causality in the real-world, beyond a physics focus.

Everyday environments are rich in tangible control interfaces (TCIs), like, light switches, appliance panels, and embedded GUIs, that are designed for humans and demand commonsense and physics reasoning, but also causal prediction and outcome verification in time and space (e.g., delayed heating, remote lights).

SWITCH: Benchmarking Modeling and Handling of Tangible Interfaces in Long-horizon Embodied Scenarios (https://huggingface.co/papers/2511.17649)

Feedback, suggestions, and collaborators are very welcome!

I'm hiring and can't find enough good CS candidates. Not in the UK though.

For a while I shared an old apartment with 3 others. Old building, 10th floor, cloth-wrapped wiring (no modern plastic/rubber/etc.), windows that didn't close, a condemned fenced-off balcony, and occasional rat visitors that could reach the 10th floor.

Dupe. Recent thread: https://news.ycombinator.com/item?id=45962654

[dupe] https://news.ycombinator.com/item?id=45809932

[dupe] https://news.ycombinator.com/item?id=45809932

Just maybe not enough.

Google has DeepMind, Microsoft has Microsoft Research, Meta has FAIR.

It’s not trivial to foster such environments, but they do still exist in different forms.

My team and I are working on embodied AI. More specifically, focusing on humanoid legged robots for long horizon tasks combining navigation and manipulation/interaction.

We've recently put a paper out in arXiv (https://arxiv.org/pdf/2503.12533) and the project page is: https://beingbeyond.github.io/being-0/

BAAI is also hiring! No fully remote positions though. :-/

Not to mention that clarifying this can lead to increased motivation to learn and discover stuff, as kids won’t think that if they are not geniuses they shouldn’t even try to create something new.

Well, there was a Canadian competitor company, and then a Canadian CEO...

The funny thing is that all FAANGs have big European subsidiaries, and specially so for their research labs.

Also, they always buy any promising new European company (and many times even keep their original offices open in Europe).

If things in Europe are so bad, I wonder why?

I’m working on embodied AI. More specifically, focusing on humanoid legged robots for long horizon tasks combining navigation and manipulation/interaction.

We just actually put a paper out in arXiv: https://arxiv.org/pdf/2503.12533 And the project page is: https://beingbeyond.github.io/being-0/

BAAI is also hiring! No fully remote positions though. :-/

I’m actually working on a demo like this. Kinda. ;-)

Hope to share the details here soon.

They already do, in a sense. The scale used here is different from the internationally used and the categories imply lesser health effects.

That really made China move. Both in blocking some social media and taking more action to improve air quality.

Vitamin A neither prevents nor cures measles.

Yes^. Otherwise most of Europe and Japan would be dead by now.

^At least to some degree.

Economic powerhouses? Sure. But forces for good? I don't think necessarily so.

Regarding the wealth thing, I'd really like to see some analysis that quantifies the risks of not having guaranteed health insurance. It is basically a gamble in the US. And even when tech companies pay for your insurance, it doesn't cover everything and can disappear at any moment if you are fired.

I'd also like to see some analysis of the impact of all the European companies that US companies acquire and integrated when they show promise. Not to mention that all the big tech companies have big offices with thousands of employees in Europe too. How much is this a wealth transfer either way?

Wow! I knew there were some issues between him and regulatory agencies, but many more than I thought: FAA, DOJ, FCC, FTC, Interior, EEOC, NLRB, EPA, etc.

When I was in there in Microsoft Research, our team was working on related efforts. But, yes, while pieces have shipped into products, MS never released a complete solution at the time.

Some links that might be of interest:

- Table semantics: https://www.microsoft.com/en-us/research/project/table-inter...

- Entity semantics (video): https://onedrive.live.com/?authkey=%21AMIdbT4yVFaw2Kk&cid=A6...

- Natural Language in Spreadsheets: https://www.microsoft.com/en-us/research/project/gridbook/

v2 is 2012 and I think some of the other books linked are more recent.

Way to exaggerate the amount by claiming full game price for all games in a bunch of random ROMs preloaded in each console. There were only 12k seized consoles...

We've built Cradle, a framework that leverages such models to perform complex computer tasks via the same general interface humans use: screen as input and keyboard & mouse operations as output.

It works both on regular software and in complex games like RDR2. And it doesn't cheat by using any game-/software-specific API, nor accessibility calls, nor DOM trees. :)

https://baai-agents.github.io/Cradle/

And we're still evolving it.

"The Woman"

Discussion: https://news.ycombinator.com/item?id=39675735

The airline itself said it was due to technical issues, right after the plane landed.

If you're willing, please share more details by email (in my profile). This can be very useful data to improve things.

Thanks for trying it out!